Investing in the securities market exposes investors to both market risk and returns. Measurement of expected returns is relatively easy since there is a generally accepted method of calculation. However, there is no consensus on the best way of quantifying security risk. The classical method is to use variance; over time, a number of alternative methods have been developed. This paper contributes to literature by examining the explanatory power of the nine most cited alternative risk measures in a comprehensive model. Empirical analysis is performed using regression analysis. The main result of the paper is the observation of a direct relationship between risk and returns, as predicted by theory. The risk measures that display consistent, significant explanatory power are Sharpe's β, kurtosis, value at risk and market capitalization. Results of the Kolmogorov-Smirnov test for normality connfirm that 82% of the security returns did not come from a normal distribution. The paper also develops the critical line algorithm used for calculating optimal portfolios leading to the derivation of the efficicient frontier.


Ramsay, John

Second Advisor

Sell, John


Economics; Mathematics


Applied Mathematics


portfolio and risk management

Publication Date


Degree Granted

Bachelor of Arts

Document Type

Senior Independent Study Thesis



© Copyright 2012 Nancy Tinoza